Medication safety in older adults in India : polypharmacy, self-medication, deprescribing, and informal care provider interventions

Background: Older adults (aged >60 years) are susceptible to unsafe medication practices, such as polypharmacy (taking five or more daily medicines) and self-medication (without a medical practitioner's prescription). Cardiac autonomic function commonly deteriorates with advancing age, and polypharmacy may further increase this risk. This has led to increasing attention to deprescribing as a key strategy in optimizing medication safety among older adults. From the health system point of view, in low- and middle-income countries (LMICs), informal healthcare providers (IHCPs) constitute an integral component of the healthcare system. They cater to a diverse population, including older adults, and may contribute to unsafe medication practices. Objectives: Study I investigated whether polypharmacy is linked to changes in cardiovascular autonomic function among older adults with chronic illnesses. Study II assessed the prevalence and determinants of polypharmacy and self- medication among older adults. Study III appraised the literature on the effect of deprescribing preventive medications compared to continuation on clinical, functional, and patient-centered outcomes among older adults with advanced frailty, dementia, or limited life expectancy. Study IV assessed the literature on interventions targeting IHCPs to improve healthcare services in LMICs. Methods: Study I employed a record-based cross-sectional design among older adults attending a health consultation clinic in Kolkata, India. The anticholinergic burden (ACB) and key autonomic outcomes were measured by Valsalva ratio (maximum heart rate during a Valsalva maneuver: minimum heart rate 30 s later), orthostatic hypotension, rise in diastolic blood pressure after an isometric hand- grip exercise, and heart rate variability (heart rate during expiration: heart rate during inspiration). The main outcome was the proportion of participants diagnosed with cardiac autonomic neuropathy (CAN). Binary logistic regression was employed to examine the association between polypharmacy, ACB, and CAN. Study II was a cross-sectional study conducted among 600 older adults residing in six Indian cities. Interviews were conducted in the households using a pre-tested questionnaire. Potentially inappropriate medications (PIMs) and potential prescribing omissions (PPOs) were identified using the updated STOPP-START (Screening Tool of Older Persons' Prescriptions and Screening Tool to Alert to Right Treatment) criteria, respectively. The knowledge, attitude, and reported practice (KAP) regarding self-medication were assessed using a questionnaire. The major outcome was the prevalence of polypharmacy. Binary logistic regression was used to identify factors associated with polypharmacy and self-medication. In Study III (systematic review and meta-analysis), PubMed, Embase, Cochrane Library, Web of Science, CINAHL, and ProQuest Dissertations & Theses Global databases were searched for studies that included deprescribing, tapering, or de-intensification of preventive medications (antihypertensives, statins, anticoagulants, and antidiabetics) among older adults. The primary outcome was all-cause mortality. Secondary outcomes were hospitalization, major adverse cardiovascular events (MACE), changes in blood pressure, and quality of life (QoL). Data were pooled [relative risk (RR), mean difference, or standardized mean difference] using random-effects models. The evidence certainty was evaluated by the GRADE (Grading of Recommendations Assessment, Development and Evaluation) framework (PROSPERO ID: CRD420251147086). In Study IV (systematic review), PubMed, Cochrane CENTRAL, and Embase databases were searched for studies evaluating interventions aimed at improving healthcare delivery by IHCPs in LMICs. Outcomes included changes in KAP related to case diagnosis and management, contraceptive use, referral services, and medication use. The evidence certainty was evaluated by the GRADE framework (PROSPERO ID: CRD42024521739). Results: In Study I, of 305 eligible older adults (mean age 68.9+3.4 years; 65.9% male), 81 (26.6%) were receiving polypharmacy. Within this group, 42 individuals had 98 documented PIMs. Compared to individuals not on polypharmacy, those on polypharmacy showed significantly higher cumulative ACB and a higher proportion with CAN (8.1+2.3 vs. 2.3+0.9; p=0.03 and 56.8% vs. 44.6%; p=0.01). After adjustment, polypharmacy and a cumulative ACB of >3 were both associated with greater odds of CAN [adjusted odds ratio (aOR): 2.66 (95% confidence interval (CI): 0.91-3.98) and aOR: 2.51 (95% CI: 0.99-3.52), respectively]. In Study II, the polypharmacy prevalence was 33.7% (95% CI: 29.9- 37.6%) and was significantly associated with three or more comorbidities [aOR: 2.5 (95% CI: 1.1-4.1)], hospitalization within the last three months [aOR: 4.6 (95% CI: 2-7.7)], and care transitions within the last three months [aOR: 3.3 (95% CI:1.4- 5.7)]. Prescriptions containing PIMs and PPOs accounted for 28.8% (95% CI: 25.2-32.6%) and 20.3% (95% CI: 17.2-23.8%) of individuals, respectively. The prevalence of self-medication was 19.7% (95% CI: 16.6-23.1%) and was independently associated with three or more comorbidities [aOR: 3.0 (95% CI: 1.4-6.7)], hospitalization within the last three months [aOR: 4.8 (95% CI: 1.5-8.0)], and living alone [aOR: 4.5 (95% CI: 2.4-6.6)]. Among individuals who practiced self-medication, 65.3% had limited knowledge about it, and 50% were unaware of the potential risks; 40.7% indicated engaging in unsafe self-medication behaviors. In Study III, from 10,397 records, 15 studies (>33,000 participants) were included. Overall, deprescribing was not associated with increased risk of all-cause mortality [RR: 1.15 (95% CI: 0.98-1.35), I2 : 93%], hospitalization [RR: 0.93 (95% CI: 0.82-1.07), I2 : 68%], or MACE [RR: 1.37 (95% CI: 0.70-2.70), I2 : 95%] (certainty: very low GRADE). Deprescribing was also not associated with increased risks of fracture, fall, or deterioration of QoL, but with slightly increased systolic blood pressure (deprescribing antihypertensives). In Study IV, 38 studies were included from a total of 7,255 screened studies. The trained IHCPs included community health workers, traditional healers, birth attendants providing community-level support, and medicine sellers. Training-based educational programs were the main intervention used along with healthcare services, community-based interventions, and policy and guidelines. Most interventions were multifaceted. The diseases and service area largely focused on child and maternal health, common infectious diseases, sexually transmitted diseases/infections, contraception, and medication use. Improvements were observed in diagnostic and case management skills, KAP, referral systems, contraceptive use, and appropriate medication use (certainty: very low GRADE). Approximately 25% of studies indicated negative results. Conclusion: This thesis provides an integrated evidence base to strengthen medication safety among older adults in India by linking community-level medication risks, clinical optimization strategies, and health-system realities. Study I revealed that polypharmacy was significantly associated with CAN in community-dwelling older adults. Study II demonstrated a substantial burden of polypharmacy, PIMs, PPOs, and self-medication, with risks concentrated among older adults with multimorbidity, recent hospitalization, and care transitions, highlighting critical points for medication review and reconciliation. Study III showed that deprescribing preventive medications in frail or end-of-life older adults did not increase mortality, hospitalization, or MACE, supporting deprescribing as a safe and patient-centered approach when aligned with goals of care and supported by monitoring. Finally, Study IV showed that trainingbased and multifaceted approaches can improve care practices by IHCPs in LMICs, although effects are inconsistent and evidence certainty is very low. The findings support a continuum model for safer medication use focusing on older adults. It integrates evidence on CAN, polypharmacy, deprescribing, and the role of IHCPs to develop a future comprehensive, community-based intervention package for improving medication safety among older adults in LMIC settings. Keywords: Anticholinergic burden (ACB), autonomic function, cardiac autonomic neuropathy (CAN), deprescribing, frailty, healthcare services, informal healthcare providers (IHCPs), low-and middle-income countries (LMICs), older adults, polypharmacy, potentially inappropriate medications (PIMs), potential prescribing omissions (PPOs), preventive medications, Screening Tool of Older Persons’ Prescriptions and Screening Tool to Alert to Right Treatment (STOPPSTART), self-medication. List of scientific papers I Samajdar SS, Das S, Sarkar S, Mukherjee S, Pathak A, Stålsby Lundborg C, Saha I, Tripathi SK, Pal J, Chatterjee N, Joshi SR. Association between Polypharmacy and Cardiovascular Autonomic Function among Elderly Patients in an Urban Municipality Area of Kolkata, India: A Record-Based Cross-Sectional Study. Geriatrics (Basel). 2022;7(6):136. https://doi.org/10.3390/geriatrics7060136 II Das S, Gnanavel P, Smanla S, Saikia AM, Mishra S, Khare S, Murugan SA, Prakash VN, Roychoudhury P, Ahmad I, Diwan V, Rajesh J, Kumar KS, Kishore J, Srivastava N, Yasmin S, Nasrin M, Borah R, Basistha MS, Baruah C, Kalita M, Samajdar SS, Cherian JJ,